A major obstacle for HIV-1 vaccine development is virus diversity which continues to increase due to high mutation rates and recombination. Within a patient, HIV-1 rapidly evolves and effectively stays one step ahead of the acquired host immune response. Although most patients have robust humoral response with HIV-1 neutralizing antibodies (Nabs) as well as HIV-1 specific CTL response, attempts to generate immunogens capable of eliciting broadly reactive Nabs targeting conserved regions of the envelope have been largely unsuccessful. However, we believe that even though HIV-1 evolution is continuous and results in an accumulation of mutations in hypervariable regions (e.g. env), there are still a finite number of genotypes that are necessary for functional viral proteins and infectivity. Therefore, theoretically, a pool of Abs targeting variable regions of all (or most) HIV-1 isolates should have broadly neutralizing activity to overcome this viral diversity. Here we are going to develop and test a polyvalent HIV-1 vaccine with increasing numbers of immunogens (5, 50, and 500) that represent functional Env glycoproteins. We have developed a unique intersubtype recombination methodology to generate the full array of HIV-1 envelopes. This technique only produces functional envelopes and differs from random mutagenesis that the vast majority of env genes would produce non-functional env proteins. We have also designed a three vector system to produce SHIVenvrec and HIV- 1envrec vaccine vectors that can infect a cell, integrate into host DNA, and express envelope but is incapable of virus propagation or spread. The modified non-infectious vaccine vectors will elicit both humoral and cellular immune response. In Aim 1, we will generate various HIV-1 A/D envelope recombinants through our unique intersubtype recombination technique, map the recombination breakpoints, and select ~500 A/D envelope recombinants to generate SHIVenvA/D vaccine vectors for vaccine studies. In Aim 2, we will investigate humoral response in different mouse models immunized with heterogeneous envA/D recombinants-based vaccine and screen for broadly neutralizing anti-HIV antibodies. In Aim 3, we will test humoral and cellular responses in Rhesus Macaques, and find out the breadth and efficiency of protection induced by these envA/D polyvalent vaccines. We expect two possible outcomes: (1) a vaccine containing heterogeneous env genes may elicit a broad HIV-specific immune response to all presented antigens from this heterogeneous immunogen, (2) a highly variable env immunogen, with regions of hypervariability, may result in "interference", disrupting the recognition of the variable sites, and may lead to a focused response on more conserved regions of the immunogen. If the first point is more evident, future studies can involve the recombination of different HIV-1 subtype using the same techniques described in this proposal. If the second point is more evident, there may be little need to provide a more diverse immunogen (than already tested in this proposal) but the inclusion of more unique intersubtype env clones into a vaccine may further focus the immune response to conserved regions. The major obstacle for HIV vaccine development is the diversity of the virus due to its high rate of mutation and recombination. Use of multivalent vaccines containing genetic or peptide material from many different strains is a strategy to overcome this issue. In this proposal, we have designed a strategy to generate intersubtype envelope recombinants-based polyvalent SHIVenv vaccines that might induce broadly effective immune response in Rhesus macaques, which could be applied for future human use.